Trouble with SRA toolkit fastq-dump

[Susanna5]

Susanna5:

In your $HOME (directory you land-in after logging in) there should be a directory called ".ncbi". (NOTE: The "." you will not see this directory unless you use "ls -a" option).

You could try to delete that directory and try to re-run the perl script again.

I do not want to make this complicated but is your $HOME directory located on a partition that is mounted using "NFS" (it appears to be). If you do not know what this is then do not worry. You can check this with your system admin/someone who may know.

We had determined (working with SRA support) that the perl script/java option did not properly work when the $HOME was mounted via NFS (at least that was the only logical explanation). At this time the only option seems to copy (or edit) a working config file into the ".ncbi" directory and adjust paths accordingly. SRA warns you not to edit the file manually so I hesitate to post a version here.

If your $HOME is located on local disks then the configure script works properly.

Thanks very much for responding, and so quickly! Your suspicion is correct, my HOME is mounted via NFS. So, if I understand this correctly, there is no way to download data from the SRA for my situation? I am running windows actually, but from the SRAtoolkit guide, I understood that it was only for linux and Amazon Cloud. I'm a bit confused about your other comment over windows, but perhaps that is with NCBI Decryption Tools?

[GenoMax]

Susanna5:

In your $HOME (directory you land-in after logging in) there should be a directory called ".ncbi". (NOTE: The "." you will not see this directory unless you use "ls -a" option).

You could try to delete that directory and try to re-run the perl script again.

I do not want to make this complicated but is your $HOME directory located on a partition that is mounted using "NFS" (it appears to be). If you do not know what this is then do not worry. You can check this with your system admin/someone who may know.

We had determined (working with SRA support) that the perl script/java option did not properly work when the $HOME was mounted via NFS (at least that was the only logical explanation). At this time the only option seems to copy (or edit) a working config file into the ".ncbi" directory and adjust paths accordingly. SRA warns you not to edit the file manually so I hesitate to post a version here.

If your $HOME is located on local disks then the configure script works properly.

[Susanna5]

I'm having the exact same issue right now.

I don't have a preferences file in the /bin/ncbi directory

Same here. If running on unix, I get the following procedure:

> perl configuration-assistant.perl
==========================================
Welcome to the SRA Toolkit Configuration Script.
SRA toolkit documentation:

http://www.ncbi.nlm.nih.gov/Traces/sra/std

==========================================

cwd = '/data/home/svdven/sratoolkit.2.3.2-centos_linux64/bin'

checking for fastq-dump (local build)... no
checking for fastq-dump (./fastq-dump: found)... yes
checking for sam-dump (local build)... no
checking for sam-dump (./sam-dump: found)... yes
checking for vdb-config (local build)... no
checking for vdb-config (./vdb-config: found)... yes

Reading configuration
refseq/servers: not found
refseq/volumes: not found
refseq/paths: not found

repository: found
repository/site: not found

repository/remote: found
repository/remote/protected/CGI/resolver-cgi: 'http://www.ncbi.nlm.nih.gov/Trace s/names/names.cgi'
repository/remote/main/NCBI/apps/refseq/volumes/refseq: 'refseq'
repository/remote/main/NCBI/apps/wgs/volumes/fuseWGS: 'wgs'
repository/remote/main/NCBI/apps/sra/volumes/fuse1000: 'sra-instant/reads/ByRun/ sra'
repository/remote/main/NCBI/disabled: not found
repository/remote/main/NCBI/root: 'http://ftp-trace.ncbi.nlm.nih.gov/sra'
repository/user: found
repository/user/main/public/apps/refseq/volumes/refseq: 'refseq'
repository/user/main/public/root: '/data/home//svdven/ncbi/public': don't exist
repository/user/main/public/cache-enabled: 'true'
repository/user/main/public/apps/sra/volumes/sraFlat: 'sra'
repository/user/main/public/apps/wgs/volumes/wgsFlat: 'wgs'

Configuration is incorrect
Your configuration is incomplete. Would you like to fix it? [Y/n] Y

Fixing configuration

Would you like to enable Remote Internet access to NCBI(recommented)? [Y/n] Y
'repository/remote/main/NCBI/disabled' => 'no': updated


Checking configuration
refseq/servers: not found
refseq/volumes: not found
refseq/paths: not found

repository: found
repository/site: not found

repository/remote: found
repository/remote/protected/CGI/resolver-cgi: 'http://www.ncbi.nlm.nih.gov/Traces/names/names.cgi'
repository/remote/main/NCBI/apps/refseq/volumes/refseq: 'refseq'
repository/remote/main/NCBI/apps/wgs/volumes/fuseWGS: 'wgs'
repository/remote/main/NCBI/apps/sra/volumes/fuse1000: 'sra-instant/reads/ByRun/sra'
repository/remote/main/NCBI/disabled: 'no'
repository/remote/main/NCBI/root: 'http://ftp-trace.ncbi.nlm.nih.gov/sra'
repository/user: found
repository/user/main/public/apps/refseq/volumes/refseq: 'refseq'
repository/user/main/public/root: '/data/home//svdven/ncbi/public': don't exist
repository/user/main/public/cache-enabled: 'true'
repository/user/main/public/apps/sra/volumes/sraFlat: 'sra'
repository/user/main/public/apps/wgs/volumes/wgsFlat: 'wgs'

Configuration is correct

Then checking the directory of ncbi, I don't see any preferences. I'm following the guide on the toolkit as I'm new to this. Any help would be appreciated as I'd like to get started on the downloading of data from the SRA, and keep getting caught in this sort of problems.

P.s. I also did the java option.. java -jar sratoolkit.jar

[GenoMax]

On Windows you should do the following (once):

Run the "sratoolkit.jar" file located in "sratoolkit.2.3.2-4-win64\bin" directory. This file should be identified as "executable jar", if you have java installed.

As a part of the installation, the program will ask you to identify a location to store reference files downloaded from NCBI/SRA. The default location is "c:\users\your_windows_user_name\ncbi\public". Make changes (if needed) and allow the program to complete the configuration.

Once the configuration is complete you should see a configuration file ("user-settings.mkfg") created in c:\users\your_windows_user_name\.ncbi". NOTE: The config files in "bin\ncbi" directory are part of the installer.

Now you should be able to open a command window and run "Provide_PATH_TO\fastq-dump SRR#.sra". Program will automatically determine if the SRA file is based on reference compression and will download the necessary reference files from NCBI/SRA (and store them in the location chosen above).

[oaklander141]

I'm having the exact same issue right now.

I don't have a preferences file in the /bin/ncbi directory

[GenoMax]

Thanks for the suggestion.

I am using the latest version 2.3.2. I ran the perl script as suggested and wasn't prompted with any questions as your post from the March thread suggested. The script ran and then read "Configuration is correct"

I'm by no means a bioinformatics expert, but I thought downloading the SRA toolkit and running fastq-dump should be straightforward.

What OS are you doing this on?

If this is on unix then look at the ".ncbi" directory in your $HOME. There should be a preferences file in it that should have several lines. If it doesn't then that means the script did not work correctly. This setup only has to be done once per user but if it does not work correctly the first time (it does not explicitly tell you if it does not) then you will see the error you reported.

SRA is switching to reference based compression for .sra files. If the sra file you are looking to expand is one of the newer ones then you have to run the perl script or the java jar to setup a location for storage of the reference files that are downloaded from SRA on the fly.

[ALB_1]

Thanks for the suggestion.

I am using the latest version 2.3.2. I ran the perl script as suggested and wasn't prompted with any questions as your post from the March thread suggested. The script ran and then read "Configuration is correct"

I'm by no means a bioinformatics expert, but I thought downloading the SRA toolkit and running fastq-dump should be straightforward.

[GenoMax]

See post#6 in this thread. I assume you are using the newest (v.2.3.2) sratoolkit.

[ALB_1]

Hi there, I am trying to use fastq-dump on an sra file downloaded from NCBI GEO and keep getting the error message:

err: name not found while resolving tree within virtual file system module - failed to open 'SRRfilenamehere'
Written 0 spots total

I downloaded the correct version and have used this toolkit successfully on other computers. Could it be an installation issue? Thank you!